Dynamic virtual generator-based specific bus-connected power system inertia resource modeling method and system

Abstract

Provided, according to the present embodiments, is a dynamic virtual generator-based specific bus-connected power system inertia resource modeling method comprising the steps of: a virtual generator generating response power from a plurality of inertia resources connected to a specific bus, and transmitting the response power to a power system; collecting the response power as set unit power and separating the response power of individual inertia resources; comparing the power system response power with a model response power calculated in a dynamic virtual generator model, and if a mismatch occurs therebetween, correcting the model response power and analyzing time response characteristics of the corrected model response power, thereby estimating an inertia constant and a damping constant; and verifying consistency by comparing the inertia constant and the damping constant estimated in the dynamic virtual generator model with actually measured response power data measured in the power system.

Description

Dynamic Virtual Generator-Based Inertia Resource Modeling Method and System for Specific Busbar-Linked Power Systems

[0001] The present disclosure relates to a dynamic virtual generator-based inertia resource modeling technique for a specific busbar-linked power system.

[0002] Generally, power system inertia acts as a key factor in maintaining frequency stability, as the rotating mass of generators plays a role in mitigating system frequency fluctuations. However, with the recent increase in the proportion of new and renewable energy sources, the ratio of synchronous generators with mechanical rotating mass has decreased, and the proliferation of power electronics-based inverter interconnection facilities has led to a problem where the level of effective inertia of the entire system is declining.

[0003] Under these circumstances, system operators need a model that estimates the inertia constant (H) and damping constant (D) of the actual system to determine the system's frequency response characteristics in real time and evaluates system stability based on this.

[0004] However, conventional inertia estimation techniques have been implemented by ignoring damping constants (D) or estimating only inertia constants (H) using only some simplified parameters. In this case, since damping factors are not considered, the actual damping behavior of the frequency response is not accurately reflected, and consequently, there is a limitation in that the precision of the inertia response model is reduced.

[0005] Furthermore, conventional virtual inertial generator models simplified the representation of inertial response characteristics, resulting in the inclusion of numerous unnecessary parameters and making it difficult to accurately simulate the effective characteristics of the inertial response. In particular, the absence of mechanisms to correct response errors between the actual system and the model, or to dynamically estimate inertial and damping constants, led to reduced reliability in inertial response modeling and decreased accuracy in system stability assessments.

[0006] Therefore, a new modeling technique is required to improve the accuracy of inertial response dynamics, including damping constants, and to calculate accurate inertial response characteristics by correcting the difference between the actual system response and the model response in real time.

[0007] These embodiments can provide dynamic virtual generator-based specific busbar-linked power system inertia resource modeling technology.

[0008] In one aspect, the present embodiments may provide a dynamic virtual generator-based inertia resource modeling method for a power system connected to a specific bus, comprising the steps of: generating response power from a plurality of inertia resources connected to a specific bus and transmitting it to the power system; aggregating the response power into a unit power and distinguishing the response power of individual inertia resources; comparing the power system response power with the model response power calculated in the dynamic virtual generator model, correcting the model response power if a discrepancy occurs, and estimating inertia constants and damping constants by analyzing the time response characteristics of the corrected model response power; and verifying consistency by comparing the inertia constants and damping constants estimated in the dynamic virtual generator model with actual response power data measured in the power system.

[0009] In another aspect, the present embodiments may provide a dynamic virtual generator-based power system inertia resource modeling system for a specific bus-linked power system, comprising: a virtual generator that generates response power from a plurality of inertia resources linked to a specific bus and transmits it to the power system; a first grid that aggregates the response power output from the virtual generator into a collective unit power and distinguishes the response power of individual inertia resources; a dynamic virtual generator model that compares the power system response power input from the first grid with the model response power calculated in the model, corrects the model response power when a discrepancy occurs, and analyzes the time response characteristics of the corrected model response power to estimate inertia constants and damping constants; and a second grid that verifies consistency by comparing the inertia constants and damping constants estimated by the dynamic virtual generator model with actual response power data measured in the power system.

[0010] According to the embodiments of the present disclosure, inertia response dynamic characteristics including damping constants can be precisely estimated based on response power obtained from a plurality of inertia resources, and inertia constants at the individual and collective levels can be accurately calculated using the time response characteristics of the corrected model response power. Furthermore, by comparing and verifying the measured response power waveform and the model response power waveform according to a reference value, absolute and relative evaluation of inertia characteristics by time period or system configuration is possible, thereby providing reliable inertia response analysis and system stability diagnosis results even in an inertia degradation environment.

[0011] FIG. 1 is a system configuration diagram of a dynamic virtual generator according to one embodiment.

[0012] Figure 2 is a diagram exemplarily showing the inertia response characteristic transfer function applied in the dynamic virtual generator model of Figure 1.

[0013] FIG. 3 is a flowchart illustrating the parameter estimation procedure of a virtual inertia model according to one embodiment.

[0014] FIG. 4 is a diagram exemplarily showing a test system for verifying a virtual generator inertia model according to one embodiment.

[0015] Hereinafter, some embodiments of the present disclosure will be described in detail with reference to the exemplary drawings. In assigning reference numerals to the components of each drawing, the same components may have the same reference numeral as much as possible, even if they are shown in different drawings. Furthermore, in describing the embodiments, if it is determined that a detailed description of related known components or functions may obscure the essence of the technical concept, such detailed description may be omitted. Where terms such as "comprising," "having," or "consisting of" are used in this specification, other parts may be added unless "only" is used. Where a component is expressed in the singular, it may include a plural unless otherwise specified.

[0016] Additionally, terms such as first, second, A, B, (a), (b), etc., may be used to describe the components of the present disclosure. These terms are used merely to distinguish the components from other components, and the nature, order, sequence, or number of the components are not limited by such terms.

[0017] In describing the positional relationship of components, where it is stated that two or more components are "connected," "combined," or "joined," it should be understood that while the two or more components may be directly "connected," "combined," or "joined," they may also be "connected," "combined," or "joined" with other components "intervened." Here, the other components may be included in one or more of the two or more components that are "connected," "combined," or "joined" with one another.

[0018] In describing the temporal flow relationship regarding components, methods of operation, or methods of production, for example, when the temporal or sequential relationship is described using "after," "following," "next," or "before," it may include cases where the relationship is not continuous unless "immediately" or "directly" is used.

[0019] Meanwhile, where numerical values ​​or corresponding information regarding a component (e.g., levels, etc.) are mentioned, even without separate explicit notation, the numerical values ​​or corresponding information may be interpreted as including a range of error that may occur due to various factors (e.g., process factors, internal or external shocks, noise, etc.).

[0020] FIG. 1 is a system configuration diagram of a Dynamic Virtual Power Plant (DVPP) according to one embodiment.

[0021] As illustrated in FIG. 1, the dynamic virtual generator (100) may include various flexible resources such as a fuel cell (FC) (110), a grid-forming energy storage system (ESS GFL) (120), a grid-following energy storage system (ESS GFM) (130), a vehicle-to-grid (V2G) (140), a power-to-hydrogen converter (P2H) (150), and a synchronous condenser (160).

[0022] The output generated from these flexibility resources can first flow into the dynamic virtual generator (DVPP) system via the first data input point (200). Here, the first data input point (200) can serve as an entry point for the response power of the individual resources to enter the dynamic virtual generator system.

[0023] The incoming output can be collected and combined at the first combination node (300), and the collected power can be connected to the grid through the first grid connection point (400). Subsequently, at the first grid (500), the output response of each resource is used for model analysis of the individual response power (Individual P DVPP It can be transmitted in the form of Individual Response Power.

[0024] Individual response power can be transmitted to a Dynamic Virtual Power Plant model (DVPP) (600). The Dynamic Virtual Power Plant model (600) first converts the input response power of the Dynamic Virtual Power Plant (100) into a set unit model, and can correct the difference between the actual response power data and the model output by comparing them during an analysis process based on an Ensemble Kalman Filter (ENKF). Subsequently, in the prediction process, the inertia constant (H) and damping constant (D) can be estimated based on the corrected state.

[0025] At this time, the dynamic virtual generator model (600) collectively sums the inertia constants (H) of individual resources to obtain an equivalent inertia constant (H DVPP ) can be calculated. This can be defined by the following mathematical formula 1.

[0026] [Mathematical Formula 1]

[0027]

[0028] J: Rotational inertia

[0029] ω : Angular velocity of the system

[0030] Srated: Rated capacity of the system

[0031] H i : Inertia constant of individual resources (fuel cells, grid-type energy storage systems, grid-following energy storage systems, electric vehicles, power-to-hydrogen converters, synchronous condensers, etc.)

[0032] That is, in one embodiment, the inertia contribution of individual equipment is collectively summed to obtain the equivalent inertia constant (H) of the dynamic virtual generator model (600). DVPP ) can be calculated, and based on this, a virtual inertia model that matches the actual system response can be implemented.

[0033] The result derived through this process is the estimated response power (Estimated P DVPP It can be calculated as ) and can be output externally through the second data output point (700). The second data output point (700) can serve as an outlet before the internal result of the dynamic virtual generator model (600) goes out to the external system.

[0034] The output estimated response power can be collected by first passing through the second coupling node (800), then connected to the grid through the second grid linkage point (900), and finally transmitted to the second grid (1000). At the second grid (1000), a consistency verification can be performed by comparing the estimated response curve of the dynamic virtual generator model (600) with the response curve measured in the actual grid. If the result of the consistency verification satisfies the criteria, the virtual inertia modeling based on the dynamic virtual generator can be finally completed.

[0035] In addition, parameters (H) estimated through auxiliary software (PSS / e, Matlab, PSCAD, etc.) DVPP , D DVPP The validity of ) can be cross-validated, and through this, system operators can obtain a real-time inertial response model that can collectively control various flexibility resources.

[0036] Inertia response modeling based on such dynamic virtual generator models can precisely reflect the dynamic characteristics of the entire grid by integrating the characteristics of individual flexibility resources into a set. In particular, it enables the effective calculation of critical inertia levels, thereby enhancing the reliability of grid stability assessments. Furthermore, based on real-time estimated inertia constants (H) and damping constants (D), it allows for stable operation even in the event of rapid frequency fluctuations or fault situations. Additionally, it can contribute to resolving the issue of grid flexibility shortages resulting from the expansion of renewable energy and the electronization of power systems.

[0037] Figure 2 is a diagram exemplarily showing the inertia response characteristic transfer function applied in the dynamic virtual generator model of Figure 1.

[0038] As illustrated in FIG. 2, the frequency deviation (Δω) generated in the system is input to a speed-regulation droop (610) to produce an output power correction control signal (ΔPs). The speed-regulation droop (610) is in the form of a sign-inverted integrator gain block (-K i / s) and droop factor (1 / R p It can be composed of ) and can perform an initial control operation to correct for speed fluctuations in the system.

[0039] Next, at the first summing node (620), a control signal (ΔPs) output from the speed control unit (610) is input, and the discrepancy between the generator output and the grid demand can be calculated. This result can be transmitted to the Turbine-Governor Dynamics Unit (630). Here, the Turbine-Governor Dynamics Unit (630) uses a first-order delay function (1 / (T g It can be implemented in the form of s + 1)) and can model the response delay of actual mechanical turbines and governors. Through this process, the mechanical input power (ΔPm) can be calculated.

[0040] Next, at the second summing node (640), the mechanical input power (ΔPm) and the electrical output power (ΔPe) can be compared and summed. This difference is input to the Generator Dynamics Unit (650), and a frequency response reflecting the inertia constant (H) and damping constant (D) can be calculated. Here, the Generator Dynamics Unit (650) can be composed of a transfer function of the form (1 / (Ms + D)), and the frequency response (Δω) can be calculated by considering the inertia constant (H) and damping constant (D) of the generator. That is, the rotational inertia and internal damping characteristics of the generator can mathematically express how much frequency fluctuation is absorbed and stabilized.

[0041] Accordingly, FIG. 2 shows the generator inertia response to system frequency fluctuations in stages through a structure leading from the speed control unit (610) → the first summing node (620) → the turbine-governor dynamic characteristic unit (630) → the second summing node (640) → the generator dynamic characteristic unit (650). In one embodiment, by applying an ensemble Kalman filter (ENKF) algorithm based on this transfer function to precisely estimate the inertia constant (H) and the damping constant (D), an improved dynamic virtual generator inertia response model compared to existing technology can be implemented.

[0042] FIG. 3 is a flowchart illustrating the parameter estimation procedure of a virtual inertia model according to one embodiment.

[0043] First, when a perturbation signal is applied to the system (Perturbation Signal, S310), active power (P from individual equipment) G,O , P G,dt ), inverter output power (P inv,0 , P inv,dt Detailed measurement data such as the rate of change of frequency (RoCoF) can be collected (S320). This step may be a process of securing basic data to accurately reflect the dynamic characteristics occurring in the actual system.

[0044] Next, the collected data can be processed through the Ensemble Kalman Filter (ENKF) algorithm, through which the inertia constant (H) and damping constant (D), which are key parameters of the Virtual Generator Model (DVPP Model), can be estimated (S330). The Ensemble Kalman Filter algorithm has strengths in estimating state variables from noisy observations and can derive accurate parameters by gradually reducing the error through iterative updates.

[0045] At the same time, the voltage and frequency obtained from the grid can be input into the Playback Generator model (S340). Subsequently, the inertia constant (H) and damping constant (D) estimated based on the ensemble Kalman filter can be input into the GENCLS Model (S350). Through this process, the actual measurement data and the model simulation results can be aligned so that they can be compared under the same conditions.

[0046] Next, the response accuracy of active power (P) and reactive power (Q) can be verified by comparing the simulation results of the generator model (GENCLS Model) with the actual measured data to determine whether it satisfies the pre-set criteria (S360). At this time, if the accuracy is below the criteria, the process can return to the parameter estimation step (S330) to perform iterative correction. Conversely, if the accuracy is determined to be within the criteria, the virtual inertia modeling is finally completed and the procedure can be terminated.

[0047] This procedure goes beyond simple parameter derivation to enable the acquisition of a precise virtual inertia model that aligns with the dynamic characteristics of the actual system. In particular, by simultaneously estimating the inertia constant (H) and the damping constant (D), it is possible to incorporate damping effects that were previously ignored by existing technologies, thereby enabling more reliable system stability assessment and critical inertia calculation.

[0048] Furthermore, the virtual inertia modeling technique according to the present embodiment does not merely estimate inertia constants (H) and damping constants (D), but includes a structure that iteratively verifies and corrects the dynamic alignment between response waveforms occurring in the actual system and model-based simulation results. This provides the advantage of enabling the derivation of stable inertia parameters even in environments where system configuration and operating conditions fluctuate. This iterative verification-based model alignment procedure allows for the continuous tracking of the contributions of virtual and physical inertia, even in situations where uncertainty in system response has increased due to the growing proportion of renewable energy generation, and enables a more precise evaluation of the reference inertia level required to ensure system stability.

[0049] FIG. 4 is a diagram exemplarily showing a test system for verifying a virtual generator inertia model according to one embodiment.

[0050] As described, a model that reproduces system changes may be placed on the left side of the system, and voltage and power changes generated from this model may be applied to the test system. At this time, a first measuring device (e.g., SC_PMU, 6023) may be installed at the first bus (1100). The first measuring device can collect voltage, frequency, active power (P), and reactive power (Q) data generated within the system in real time. The data collected in this way can be utilized in the parameter estimation procedure based on the Ensemble Kalman Filter (ENKF) of FIG. 3 described above, and can serve as basic data to verify whether the estimated inertia constant (H) and damping constant (D) match the actual system dynamic characteristics.

[0051] Next, a transmission line (1200) containing impedance components (L, R, etc.) may be located in the center of the test system. This section can reproduce the loss, phase delay, and transmission characteristics of an actual transmission line, allowing the virtual generator model to be verified in an environment similar to an actual power grid.

[0052] The output passing through the transmission line (1200) can be input to a model verification unit (e.g., a synchronous condenser and inverter-based device) through a second measuring device (e.g., SC, 26203) installed at the second bus (1300) on the right. This point can serve to measure output data after the transmission line (1200), and the dynamic virtual generator (DVPP) based output response and the actual grid response can be directly compared in the model verification unit.

[0053] That is, the test system of FIG. 4 is structured to follow the system change reproduction model on the left → the first bus (1100) → the central transmission line (1200) → the second bus (1300) → the model verification section on the right. During this process, the active power (P) and reactive power (Q) response curves at each point are collected, and the results can be evaluated for consistency with actual system measurements according to accuracy criteria.

[0054] If the evaluation results satisfy the criteria, the inertia constant (H) and damping constant (D) estimated in Figures 2 and 3 above are confirmed to be consistent with the actual system dynamic characteristics, and finally, the virtual inertia modeling based on the Dynamic Virtual Generator (DVPP) can be completed. Conversely, if the criteria are not satisfied, the process can be returned to the parameter estimation step based on the Ensemble Kalman Filter (ENKF) to perform iterative correction.

[0055] Accordingly, FIG. 4 presents a test system structure that allows a virtual generator inertia model of one embodiment to be verified in an environment identical to actual system conditions, thereby enabling modeling that reflects accurate inertia response characteristics.

[0056] The foregoing description is merely an illustrative explanation of the technical concept of the present disclosure, and those skilled in the art to which the present disclosure pertains may make various modifications and variations within the scope of the essential characteristics of the technical concept. Furthermore, since these embodiments are intended to explain, not limit, the scope of the technical concept is not limited by these embodiments. The scope of protection of the present disclosure shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of the present disclosure.

[0057]

[0058] CROSS-REFERENCE TO RELATED APPLICATION

[0059] This patent application claims priority pursuant to Section 119(a) of the U.S. Patent Act (35 USC §119(a)) to Korean Patent Application No. 10-2024-0186138 filed on December 13, 2024 and Korean Patent Application No. 10-2025-0160096 filed on October 30, 2025, all of which are incorporated by reference into this patent application. Additionally, this patent application claims priority in countries other than the United States for the same reasons as above, all of which are incorporated by reference into this patent application.

Claims

1. As a dynamic virtual generator-based method for modeling inertia resources of a specific busbar-linked power system, A step in which a virtual generator generates response power from multiple inertial resources connected to a specific busbar and transmits it to the power system; A step of aggregating the above response power into aggregation unit power and distinguishing the response power of individual inertia resources; A step of comparing the power system response power with the model response power calculated in the dynamic virtual generator model, correcting the model response power if a discrepancy occurs, and analyzing the time response characteristics of the corrected model response power to estimate inertia constants and damping constants; and A dynamic virtual generator-based inertia resource modeling method for a specific bus-linked power system, comprising the step of verifying consistency by comparing the inertia constants and damping constants estimated in the above dynamic virtual generator model with actual response power data measured in the power system.

2. In Paragraph 1, The above virtual generator is, A dynamic virtual generator-based method for modeling specific busbar-linked power system inertial resources, comprising at least one of a fuel cell, a grid-type energy storage device, a grid-following energy storage device, an electric vehicle, a power-to-hydrogen converter, and a synchronous condenser.

3. In Paragraph 1, In the above estimation step, A dynamic virtual generator-based inertia resource modeling method for a specific bus-linked power system, wherein the above dynamic virtual generator model estimates the error between the power system response power and the model response power using an ensemble Kalman filter and corrects the model response power based on the error.

4. In Paragraph 3, The above ensemble Kalman filter is, A dynamic virtual generator-based specific bus-linked power system inertia resource modeling method that sets a plurality of state variable sets corresponding to power system response power collected from the power system as initial state estimates, samples the state variable sets in parallel to calculate a predicted value for each state, and then calculates the error covariance between the predicted value and the model response power to estimate the statistical distribution of the inertia constant and damping constant.

5. In Paragraph 1, In the above estimation step, The above dynamic virtual generator model is a dynamic virtual generator-based method for modeling inertia resources of a specific bus-linked power system, which calculates equivalent inertia constants by summing individual inertia constants calculated based on the rotational inertia, system angular velocity, and rated capacity of each of the plurality of inertia resources in a set unit.

6. In Paragraph 1, In the above estimation step, The above dynamic virtual generator model is a method for modeling inertia resources of a specific bus-linked power system based on a dynamic virtual generator that reflects the dynamic characteristics of the power system by applying a transfer function in the form of a first-order delay function that simulates the control delay and mechanical inertia characteristics of the virtual generator using the corrected model response power.

7. In Paragraph 1, In the above verification step, A dynamic virtual generator-based inertia resource modeling method for a specific bus-linked power system, which determines consistency by comparing the error between a model response power waveform generated using inertia constants and damping constants estimated in the dynamic virtual generator model and a measured response power waveform in the power system with a preset reference value.

8. In Paragraph 7, A dynamic virtual generator-based inertia resource modeling method for a specific bus-linked power system, wherein the active power and reactive power errors between the actual response power waveform measured in the power system and the model response power waveform generated in the dynamic virtual generator model are determined to be complete when the error is within a preset reference value range.

9. As a dynamic virtual generator-based specific busbar-linked power system inertia resource modeling system, A virtual generator that generates response power from multiple inertial resources connected to a specific busbar and transmits it to the power system; A first grid that aggregates the response power output from the above-mentioned virtual generator into aggregate unit power and distinguishes the response power of individual inertial resources; A dynamic virtual generator model that compares the power system response power input from the first grid with the model response power calculated in the model, corrects the model response power when a discrepancy occurs, and analyzes the time response characteristics of the corrected model response power to estimate inertia constants and damping constants; and A dynamic virtual generator-based specific bus-linked power system inertia resource modeling system comprising a second grid that verifies consistency by comparing the inertia constants and damping constants estimated in the above dynamic virtual generator model with actual response power data measured in the power system.

10. In Paragraph 9, The above virtual generator is, A dynamic virtual generator-based specific busbar-linked power system inertia resource modeling system comprising at least one of a fuel cell, a grid-type energy storage device, a grid-following energy storage device, an electric vehicle, a power-to-hydrogen converter, and a synchronous condenser.

11. In Paragraph 9, The above dynamic virtual generator model is, A dynamic virtual generator-based power system inertia resource modeling system connected to a specific bus, which estimates the error between the power system response power and the model response power using an ensemble Kalman filter and corrects the model response power based on the error.

12. In Paragraph 11, The above ensemble Kalman filter is, A dynamic virtual generator-based specific bus-linked power system inertia resource modeling system that sets a plurality of state variable sets corresponding to power system response power collected from the power system as initial state estimates, samples the state variable sets in parallel to calculate a predicted value for each state, and then calculates the error covariance between the predicted value and the model response power to estimate the statistical distribution of the inertia constants and damping constants.

13. In Paragraph 9, The above dynamic virtual generator model is, A dynamic virtual generator-based specific busbar-linked power system inertia resource modeling system that calculates an equivalent inertia constant by summing individual inertia constants calculated based on the rotational inertia, system angular velocity, and rated capacity of each of the plurality of inertia resources above, in a set unit.

14. In Paragraph 9, The above dynamic virtual generator model is, A dynamic virtual generator-based power system inertia resource modeling system for a specific bus-linked power system that reflects the dynamic characteristics of the power system by applying a transfer function in the form of a first-order delay function that simulates the control delay and mechanical inertia characteristics of the virtual generator using corrected model response power.

15. In Paragraph 9, The above second grid is, A dynamic virtual generator-based inertia resource modeling system for a specific bus-linked power system that determines consistency by comparing the error between a model response power waveform generated using inertia constants and damping constants estimated in the dynamic virtual generator model and a measured response power waveform in the power system with a preset reference value.

16. In Paragraph 15, The above second grid is, A dynamic virtual generator-based inertia resource modeling system for a specific bus-linked power system, which determines that virtual inertia modeling is complete when the active power and reactive power errors between the actual response power waveform measured in the power system and the model response power waveform generated in the dynamic virtual generator model exist within a preset reference value range.

Images